Copper-catalyzed domino route to natural nostoclides and analogues: A total synthesis of nostoclides i and II

Article abstract of DOI:10.1002/adsc.201300459

An original and convenient domino route to natural nostoclides I and II has been developed using a two-step sequence consisting of a copper-cat-alyzed tandem reaction associated with Suzuki cross-coupling. The methodology employed for this total synthesis appeared to be an interesting and suffi-ciently flexible tool to allow the synthesis of numer-ous analogues of these nostoclides.

Full text of DOI:10.1002/adsc.201300459

FULL PAPERS
DOI: 10.1002/adsc.201300459
Copper-Catalyzed Domino Route to Natural Nostoclides and
Analogues: A Total Synthesis of Nostoclides I and II
Samuel Inack Ngi,a,* Julien Petrignet,a Romain Duwald,a El Mostafa El Hilali,a
Mohamed Abarbri,a Alain DuchÞne,a and Jꢀrꢁme Thibonneta,*
a
Laboratoire Infectiologie Santꢀ Publique – Equipe de Recherche et Innovation en Chimie Mꢀdicinale, UMR-UFR INRA
1282, Facultꢀ des Sciences et techniques de Tours, Parc de Grandmont, 37200 Tours, France
Fax : (+33)-2-4736-7073; e-mail: samuel.inack@univ-tours.fr or inack_sam@yahoo.fr or jerome.thibonnet@univ-tours.fr
Received: May 26, 2013; Published online: October 9, 2013
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201300459.
Abstract: An original and convenient domino route ciently flexible tool to allow the synthesis of numer-
to natural nostoclides I and II has been developed ous analogues of these nostoclides.
using a two-step sequence consisting of a copper-cat-
alyzed tandem reaction associated with Suzuki cross-
coupling. The methodology employed for this total Keywords: copper catalysis; domino reactions; heter-
synthesis appeared to be an interesting and suffi- ocyclic compounds; nostoclides; total synthesis
Introduction
Argade[4c] are present in the literature. But the de-
scribed procedures differ only from the Boukouvalas
Nostoclides are metabolites extracted from Nostoc sp. route by the access to the butenolide. We report here
a blue-green cyanobacteria symbiote of the common a novel and totally different approach to access nosto-
lichen Peltigera canina. Their structure based on a g- clides I and II, exploiting a convenient method for the
butyrolactone core has been elucidated by Yang synthesis of (Z)-configured a,b-substituted g-alkylide-
et al.[1] The same authors reported that nostoclides I nebutenolides[5] in two steps including a copper-cata-
and II (Scheme 1) possess a moderate cytotoxicity to- lyzed tandem coupling–heteroannulation followed by
wards Neuro-2aCCL and KB CCL17 cancer cells, palladium-catalyzed cross-coupling.
however other works emphasize the antimicrobial,[1,2]
phytogrowth and photosynthesis[3] inhibiting activities
of nostoclide analogues.
Results and Discussion
Only a few papers deal with the synthesis of these
structures.[4] The first total syntheses of nostoclides I The reaction sequence we planned to follow is pre-
and II were reported by Boukouvalas and co-work- sented in Scheme 2. In the presence of a catalytic
ACHTUNGTRENNUNG
ers.[4a] The route proceeds via the obtaining of a suita- amount of copper iodide (CuI), terminal alkynes 4
ble g-butyrolactone before the creation of the exocy- and (E)-2,3-dihalo-4-methylpent-2-enoic acids 3 un-
clic double bond by a Mukaiyama-like condensation dergo a tandem coupling–oxacyclization reaction to
on the g-position of the obtained butenolide followed selectively yield the a-brominated nostoclides precur-
by elimination. Two other papers by Rossi[4b] and sors 2. The latter product should be able to undergo
a benzylation using the remaining a-halogen and give
the desired nostoclides. The same procedure using
palladium catalysts should fail at the first step, leading
to the b-elimination by-product.[5]
The achievement of these total syntheses involved
the efficient preparation of (E)-a,b-dihaloacrylic acid
derivatives 3 and alkynes 4. Acids 3 were synthesized
in four steps starting from isobutyraldehyde, with the
use of Corey–Fuchs[6] methodology (Scheme 3). Sub-
Scheme 1. Structures of nostoclide I and II.
mitted to the action of a mixture of triphenylphos-
2936
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 2936 – 2941

Copper-Catalyzed Domino Route to Natural Nostoclides and Analogues
Scheme 2. Retrosynthetic route to nostoclides I and II.
We initially planned to access alkynes 4 in three
steps from ortho-chlorinated and dichlorinated phe-
nols A (Scheme 4). The phenols A were firstly para-
iodinated in good yields by the action of a methanolic
mixture of sodium hydroxide, sodium iodide and
sodium hypochlorite.[8] The obtained products 8 were
reacted with trimethylsilylacetylene in a Sonogashira-
like cross-coupling.[9] The last step consisted in the re-
moval of the TMS protection by n-tetrabutylammoni-
um fluoride to afford the terminal alkynes. Alkyne 4a
was obtained from 9a in 86% yield after a fast chro-
matography because of its instability over silica gel. A
lengthy exposure to silica leads to the reduction of
the triple bond into the corresponding styrene deriva-
Scheme 3. Synthesis of (E)-a,b-dihaloacrylic acid deriva-
tives.
phine, zinc and carbon tetrabromide in equimolar tive. The recovery of the alkyne was found to be very
proportions, isobutyraldehyde afforded 1,1-dibromo- difficult by distillation. Precursor 9b was obtained in
3-methylbut-1-ene 5 in 90% yield. A treatment of 5 75% yield, then TMS was deprotected without further
with butyllithium led to the corresponding lithium purification because of its instability over silica gel.
acetylide trapped with ethyl chloroformate to afford Unfortunately the reaction led to a mixture of by-
ethyl 4-methylpent-2-ynoate 6 in 65% yield. The products and we were unable to isolate the minimal
nature of the organolithium reagent was found to be amount of alkyne obtained from it, probably because
an important factor in this step. Using n- or s-BuLi of the same instability. We tried to optimize the cleav-
promoted the formation of mixtures of by-products age of the silane by lowering the temperature or by
along with the desired compound. The use of t-BuLi adding MeLi, but these changes did not provide in-
eliminated that issue. Saponification of 6 with lithium creased yields of the alkyne.
hydroxide at room temperature provided acid 7 in
We planned an alternative route to get 4b from
90% yield. Using a previously reported[7] direct halo- protected derivatives of 10, itself obtained from 4-hy-
genation reaction of propiolic acid derivatives, acid 7 droxybenzaldehyde after treatment with N-chlorosuc-
stereoselectively led to 3a and 3b in excellent yields.
cinimide. We noticed difficulties to protect 10. Actual-
Scheme 4. Synthesis of alkynes 4. i) a MeOH, NaI, NaOH, b NaOCl. ii) Trimethylsilylacetylene, Et3N, CuI 0.1 equiv., Pd-
(PPh3)4 0.05 equiv., DMF. iii) (n-Bu)4NF, THF, À58C.
ACHTUNGTRENNUNG
Adv. Synth. Catal. 2013, 355, 2936 – 2941
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2937

Samuel Inack Ngi et al.
FULL PAPERS
Scheme 5. Attempts for the protection of dichlorophenol 10 and dichlorination of benzaldehydes 12. i) Ac2O 2.5 equiv.,
DMAP 0.05 equiv., DMF. ii) TBDMSCl 1.1 equiv., imidazole 1.1 equiv., DMF, room temperature. iii) NCS 2.5 equiv. CHCl3
reflux. iv) NCS 2.5 equiv., APTS 2.5 equiv., MeCN, room temperature. v) SO2Cl2 2 equiv., CH2Cl2.
ly in the presence of ortho-chlorines, attempts to pro- difficulties with NCS, assistance of concentrated HCl
tect the phenol with acetyl, silyl, alkyl or THP group or APTS should help to afford 15 (Scheme 6). A
afforded very poor conversion yields or large amounts treatment with t-BuLi of 15 led to alkyne 4b in good
of by-products. On the other hand treating Ac- or yield but unfortunately with styrene by-product in
TBS-protected 4-hydroxybenzaldehyde 12 with NCS a 75/25 maximum ratio in favour of alkyne. That
or SO2Cl2 was useless to obtain the corresponding di- alkyne presented more instability than 9b and 4a to-
chlorinated product 11.
wards silica gel and distillation also failed.
Attempts to chlorinate the gem-dibromo 13
In order to synthesize some analogues, we prepared
(Scheme 6) obtained from 12 via the first step of alkynes 17 in three steps from 4-methoxybenzalde-
Corey–Fuchs reaction was also a failure. Comparing hyde (Scheme 7). Even though monochlorination with
the last and previous results supported the conclusion NCS in CHCl3 or NCS APTS in MeCN[10] was easy,
that the issue for chlorination was the Ac and TBS dichlorination was very slow and led to an average
protecting groups. Actually, when protecting groups yield of 11b. Intermediates 16 were easily obtained by
were removed from 13 to give 14, chlorination with a CBr4/Zn/PPh3 treatment. Curiously, 16a was also ob-
NCS in chloroform gave the expected product. It is tained in good yield by a treatment of 15 with dimeth-
important to note that depending on the batch of yl sulfate, while the carbonyl version was unable to
NCS, the reaction may be easy or not. Some batches properly undergo this transformation (Scheme 5). Al-
were inefficient even after recrystallization. In case of kynes 17 were obtained after the treatment of 16 with
Scheme 6. Alternative synthesis of dichloroalkyne 4b. i) CBr4 2 equiv., PPh3 2 equiv. ii) Zn 2 equiv., CH2Cl2. iii) for 13a KOH
1.5 equiv., EtOH, H2O. iv) for 13b Cs2CO3 1.3 equiv., DMF, H2O. v) NCS 2.5 equiv., CHCl3. vi) a) t-BuLi 3.1 equiv., THF; b)
H2SO4.
2938
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 2936 – 2941

Copper-Catalyzed Domino Route to Natural Nostoclides and Analogues
Scheme 7. Synthesis of methoxy-protected alkynes. 1) R=H: NCS 1 equiv., R=Cl: NCS 2.5 equiv., CHCl3 reflux. ii) R=H:
NCS 1 equiv., APTS 1 equiv., R=Cl: NCS 2 equiv., APTS 2 equiv., MeCN, room temperature. iii) CBr4 2 equiv., PPh3
1
2 equiv., Zn 2 equiv., CH2Cl2. iv) a) t-BuLi 2.1 equiv., THF, b) H2SO4. Yields of 17 were evaluated via H NMR using an in-
ternal reference.
t-BuLi. The reaction yielded a mixture of alkynes and that point of view it seemed very attractive to reach
styrenes in a 93/7 maximum ratio in favour of alkyne the nostoclides precursors by using methoxy-protect-
and showed the same purification issues as the unpro- ed alkyne introducing then a deprotection sequence.
tected ones.
But our efforts to demethylate the aromatic ether
Because of issues regarding purification of alkynes from 2f–i with BBr3 or HBr in acetic acid were unsuc-
prepared from styrene derivatives which were present cessful. Studies are ongoing to find a suitable and
in all cases, we found it preferable to use the crude easy to cleave protecting group.
product directly; according to our experience the
The last step to nostoclides and analogues was
mentioned impurities do not affect the behaviour of achieved by benzylation of the position 3 of the ob-
the further step. We used then 2 equiv. of alkynes 4 tained butenolides. We tried several Pd-catalyzed
and 17, and 4-hydroxyphenylacetylene for 1 equiv. of cross-coupling methodologies. While Kumada and
acid 3 in the presence of 0.2 equiv. of CuI, according Stille coupling led to recovery of the starting material,
to the conditions described in an earlier report on the Negishi coupling led to mixtures of needed product,
copper-catalyzed orthogonal reactivity of a,b-diha- a-dehalogenated parent product and starting material
loalkenoic acid.[3] Four hours of reaction time were (Table 2, line 1). Suzuki methodology permitted us to
needed to obtain original a-halonostoclides precur- hit our goal, even if we noticed issues. It appeared in
sors 2 (Table 1) with a total stereoselectivity in favour the preliminary tests that iodinated precursors are un-
of the Z-isomer and in satisfactory isolated yields con- suitable for benzylation with a Suzuki-like procedure.
sidering the purification issues. Actually methoxy-pro- We obtained principally a-dehalogenation of the
tected analogues are more stable than free hydroxy- starting material (Table 2, lines 3–4). With the bromi-
lated and chlorinated moieties over silica gel. From nated butenolides, the reaction is more stable but
seemed to stop regardless of the source of palladium
used, its amount or the nature of the boronated reac-
tant. Actually when the reaction ran with 0.05 equiv.
Table 1. Synthesis of nostoclides precursors.
of palladium catalyst, we obtained only 45 to 55% of
conversion into the desired product even with a dou-
bled palladium load (10%) or over 24 h of reaction
(Table 2, lines 5–8).
The problem was successfully solved by a second
addition of 0.05 equiv. of palladium catalyst along
with 1 equiv. of boronated reagent 1 hour after the
first addition. The reactions gave the needed products
Entry
X
R1
R2
R3
Product
Yield [%][a]
both on methoxylated and unprotected precursors.
Slightly better yields were obtained while using the
more stable and easy to handle potassium benzyltri-
fluoroboronates[11] instead of benzylboronic acids
(Table 2, lines 10 and 11). We obtained natural nosto-
clides I and II, and several analogues in very good
yields in the last step. Considering the principal reac-
tion sequence presented in this report, these results
make the overall yields average, allowing then to
place this methodology as a serious alternative for the
synthesis of nostoclides I and II, and derivatives. The
1
2
3
4
5
6
7
8
9
Br
I
H
OH
OH
OH
H
H
H
H
Cl
H
H
Cl
Cl
H
H
H
Cl
Cl
Cl
Cl
Cl
Cl
2a
2b
2c
2d
2e
2f
2g
2h
2i
65
45
38
28
25
65
48
68
45
Br
Br
Br
Br
I
OH
OMe
OMe
OMe
OMe
Br
I
[a]
Isolated yields.
Adv. Synth. Catal. 2013, 355, 2936 – 2941
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2939

Samuel Inack Ngi et al.
FULL PAPERS
Table 2. Benzylation step of the nostoclides synthesis.
No.
Substrate
X
M
Catalyst
PdCl2A(MeCN)2
Temperature [8C]
Products obtained
Ratio[f]
Yield [%]
[a]
[c]
1
2
3
4
5
6
7
8
2b
2d
2b
2b
2c
2c
2c
2a
2a
2c
2c
2d
2e
2f
2f
2f
2h
I
Br
I
ZnBr
G
r.t.
70
70
70
70
70
80
80
80
80
70
80
80
80
80
80
80
1d/1d’/2b
1b/2d
1d/1d’
1d/1d’
1d/2c
1d/2c
1d/2c
1c/2a
1c
1d
1d
1b
1a
40/20/40
45/55
5/95
10/90
53/47
55/45
53/47
51/49
–
–
–
–
–
–
–
–
–
63
53
41
59
52
73
78
84
82
73
65
68
43
74
85
50
61
[b]
B(OH)2
B(OH)2
B(OH)2
B(OH)2
B(OH)2
BF3K
BF3K
BF3K
BF3K
B(OH)2
BF3K
Pd
Pd
U
I
PdCl
Pd
C
[b]
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
Br
A
PdCl
PdCl
PdCl
PdCl
PdCl
G
9
10
11
12
13
14
15
16
17
Pd
N
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
BF3K
BF3K
BF3K
BF3K
1e
1g
1h
1f
BF3K
[a]
BnZnBr 1.5 equiv., PdCl CHTUNGTNER(NUNG MeCN)2 0.05 equiv., DMF/THF.
2A
BnB(OH)2 1.5 equiv., Pd(PPh3)4 0.05 equiv., Na2CO3 2 equiv., toluene, EtOH.
BnB(OH)2 1.5 equiv., PdCl2A(MeCN)2 0.05 equiv., Na2CO3 2 equiv., toluene, EtOH.
BnBF3K 1.5 equiv., PdCl2A(dppf) 0.05 equiv., Na2CO3 2 equiv., toluene, H2O.
[b]
[c]
[d]
[e]
[f]
AHCTUNGTRENNUNG
CTHUNGTRENNUNG
CTHUNGTRENNUNG
A second addition of 0.05 equiv. of palladium and 1 equiv. of boron compound was made after 1 hour.
1
Ratios were measured via H NMR using an internal reference.
real advantage of this methodology comes from its Conclusions
three component configuration. It is actually possible
to envisage a rapid synthesis of numerous members of In summary we have described an original approach
the nostoclides I and II playing with the substituents to nostoclides I and II which illustrates a two-step
of the alkyne, the acid and the boronated reactant.
route for the construction of butenolide structures.
This three component methodology permits us to
vary positions 3, 4 and exocyclic substituents of the
butenolides. The transformation is sufficiently versa-
2940
asc.wiley-vch.de
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2013, 355, 2936 – 2941

Copper-Catalyzed Domino Route to Natural Nostoclides and Analogues
benzyl boron derivatives were found unsuitable for iodinat-
ed nostoclides precursors. The conditions used promoted the
a-dehalogenation of the starting material 2b into 1d’.
tile to allow the synthesis of several analogues of nos-
toclides I and II for QSAR purposes as an example.
Efforts are ongoing to properly isolate the alkynes
used and improve the recovery of nostoclides ob-
tained with our methodology.
Acknowledgements
Experimental Section
We thank the Dꢀpartement d’Analyse Chimique Biologique
et Mꢀdicale for mass analyses. Special thanks to Jean-Phil-
ippe Christidꢁs (IRBI, UMR CNRS 7261) for his great help
on early nostoclides mass analyses. Johnson Matthey Compa-
ny is thanked for generous gift of palladium metal salts.
General Procedure for the Synthesis of Nostoclides
Precursors 2
A dry Schlenk tube protected from light by an aluminium
foil and equipped with a magnetic stirrer was charged with
10.2 mmol (2 equiv.) of K2CO3, 5.0 mmol (1 equiv.) of 3 and
20 mL of DMF. The suspension was stirred for 15 min, then
the flask was evacuated cold for 10 min and backfilled with
argon. After reaching room temperature, 2 equiv. of selected
alkyne and 0.2 equiv. of CuI were added. The Schlenk tube
was placed in an oil bath pre-heated at 658C and the con-
tent stirred for 4 h. The reaction mixture was then allowed
to reach room temperature and was hydrolyzed with a satu-
rated aqueous solution of NH4Cl (20 mL). Ethyl acetate
(50 mL) was added in the Schlenk tube and the mixture was
filtered over a Celite pad. The pad was washed with addi-
tional ethyl acetate (30 mL). The aqueous phase was re-
moved and the organic layer was washed several times with
water (10ꢃ15 mL). The organic layer was dried over MgSO4
and concentrated under vacuum. The raw material obtained
was purified by flash chromatography on silica gel to give
the desired a-halo-g-alkylidenebutenolides 2.
References
[1] X. Yang, Y. Shimizu, J. R. Steiner, J. Clardy, Tetrahe-
dron Lett. 1993, 34, 761–764.
[2] a) R. R. Teixeira, L. C. A. Barbosa, C. R. A. Maltha,
M. E. Rocha, D. P. Bezerra, L. V. Costa-Lotufo, C.
Pessoa, M. O. Moraes, Molecules 2007, 12, 1101–1116;
b) L. C. A. Barbosa, C. R. A. Maltha, A. J. Demuner,
P. F. Pinheiro, J. O. S. Varejao, R. M. Montanari, N. J.
Andrade, Quꢂmica Nova 2010, 33, 2020–2026.
[3] a) L. C. A. Barbosa, A. J. Demuner, A. E. S. De, A.
Oliveira, B. King-Diaz, B. Lotina-Hennsen, Pest
Manage. Sci. 2006, 62, 214–222; b) R. R. Teixeira,
L. C. A. Barbosa, G. Forlani, D. Pilo-Veloso, J. W. d. M.
Carneiro, J. Agric. Food Chem. 2008, 56, 2321–2329.
[4] a) J. Boukouvalas, F. Maltais, N. Lachance, Tetrahedron
Lett. 1994, 35, 7897–7900; b) F. Bellina, R. Rossi, Syn-
thesis 2002, 2729–2732; c) A. Kar, S. Gogoi, N. P.
Argade, Tetrahedron 2005, 61, 5297–5302.
General Procedure for the Synthesis of Nostoclides 1
A dry Schlenk tube equipped with a magnetic stirrer and
containing H2O (2 mL) and toluene (6 mL), was charged
with 0.3 mmol of 2, 2 equiv. of potassium benzyltrifluorobor-
[5] S. I. Ngi, K. Cherry, V. Hꢀran, L. Commeiras, J.-L. Par-
rain, A. DuchÞne, M. Abarbri, J. Thibonnet, Chem.
Eur. J. 2011, 17, 13692–13696.
onate, 0.1 equiv. of PdCl2ACHTNUGTRENUNG(dppf) and 3 equiv. of Cs2CO3. The
mixture was stirred at 808C. After 1 h, additional 1 equiv. of
boronate and 0.05 equiv. of palladium were inserted and the
medium was left under stirring at 808C for 4 h. The mixture
was taken up with ethyl acetate (25 mL) and filtered
through a Celite pad. The pad was washed with additional
ethyl acetate (10 mL). The filtrate was poured in a separato-
ry funnel and the organic layer was isolated, washed with
brine (2ꢃ5 mL) and water (2ꢃ5 mL). The organic phase
was dried over MgSO4 and evaporated under vacuum. The
crude product was purified by flash chromatography on
silica gel with a suitable eluent to give the desired nostoclide
derivative 1.
[6] a) E. J. Corey, P. L. Fuchs, Tetrahedron Lett. 1972,
3769–3772; b) M. G. Organ, S. Bratovanov, Tetrahedron
Lett. 2000, 41, 6945–6949.
[7] a) S. Langle, S. I. Ngi, E. Anselmi, M. Abarbri, J. Thi-
bonnet, A. Duchene, Synthesis 2007, 1724–1728; b) S. I.
Ngi, E. Anselmi, M. Abarbri, S. Langle, A. Duchene, J.
Thibonnet, Org. Synth. 2008, 85, 231–237.
[8] K. J. Edgar, S. N. Falling, J. Org. Chem. 1990, 55, 5287–
5291.
[9] K. Sonogashira, Y. Tohda, N. Hagihara, Tetrahedron
Lett. 1975, 4467–4470.
[10] P. Bovonsombat, R. Ali, C. Khan, J. Leykajarakul, K.
Pla-on, S. Aphimanchindakul, N. Pungcharoenpong, N.
Timsuea, A. Arunrat, N. Punpongjareorn, Tetrahedron
2010, 66, 6928–6935.
The same results were obtained in the case of 2c with the
use of benzylboronic acid, PdACTHNUTRGNE(UNG PPh3)4 and aqueous Na2CO3
as the base at 708C in toluene/EtOH as solvent instead of
the corresponding reactants. Suzuki-like conditions with
[11] G. A. Molander, T. Ito, Org. Lett. 2001, 3, 393–396.
Adv. Synth. Catal. 2013, 355, 2936 – 2941
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2941